Predicting Water Resource Impacts of Unconventional Gas Using Simple Analytical Equations

Abstract

The rapid expansion in unconventional gas development over the past two decades has led to concerns over the potential impacts on groundwater resources. Although numerical models are invaluable for assessing likelihood of impacts at particular sites, simpler analytical models are also useful because they help develop hydrological understanding. Analytical approaches are also valuable for preliminary assessments and to determine where more complex models are warranted. In this article, we present simple analytical solutions that can be used to predict: (1) the spatial extent of drawdown from horizontal wells drilled into the gas-bearing formation, and rate of recovery after gas production ceases; (2) the potential for upward transport of contaminants from the gas-bearing formation to shallow aquifers during hydraulic fracturing operations when pressures in the gas-bearing formation are greatly increased; and (3) the potential downward leakage of water from shallow aquifers during depressurization of gas-bearing formations. In particular, we show that the recovery of pressure after production ceases from gas-bearing shale formations may take several hundred years, and we present critical hydraulic conductivity values for intervening aquitards, below which the impact on shallow aquifers will be negligible. The simplifying assumptions inherent in these solutions will limit their predictive accuracy for site-specific assessments, compared to numerical models that incorporate knowledge of spatial variations in formation properties and which may include processes not considered in the simpler solutions.

Towards Quantifying the Likelihood of Water Resource Impacts from Unconventional Gas Development

“…However, a recent review reports that as many as 4 to 9% of oil and gas wells have some form of observable gas leakage (Dusseault and Jackson 2014). The likelihood of groundwater contamination due to migration of fluids through an abandoned well is calculated by first estimating the minimum hydraulic conductivity (K ) of well casing that would be required to transport fluid over the required distance during the period of hydraulic fracturing, using analytical equations (Cook et al 2017). For an area with limited oil and gas development, an abandoned well density of 10 −5 km −2 seems appropriate (see Appendix), and in this case the probability of an abandoned well occurring within an assumed 1 km 2 area of influence is 10 −5 .…”

“…The average lifetime of an unconventional well is 20 to 30 years (International Energy Agency [IEA] 2009); therefore, there is as yet no observational evidence of the long-term impacts after production ceases and the likelihood of impacts must be estimated theoretically. Although during development pressure in the gas-bearing formation is reduced, after gas production ceases, pressure will increase towards its initial value, but this recovery may take several hundred years (Cook et al 2017). Although during development pressure in the gas-bearing formation is reduced, after gas production ceases, pressure will increase towards its initial value, but this recovery may take several hundred years (Cook et al 2017).…”

“…In order to develop early insight, across many systems and in a generalized, transferable, manner, it is possible to use simpler analytical approaches. Recently published analytical models (Cook et al 2017) have been developed to predict water resource impacts of unconventional gas. However, these new analytical approaches were not applied to specific sites or a more generalized likelihood of impact analysis in that paper.…”

“…However, for areas with long histories of oil and gas development, the abandoned well density might be as high as 0.1 km −2 , giving a probability of occurrence of 0.1. The likelihood of groundwater contamination due to migration of fluids through an abandoned well is calculated by first estimating the minimum hydraulic conductivity (K ) of well casing that would be required to transport fluid over the required distance during the period of hydraulic fracturing, using analytical equations (Cook et al 2017). Assuming that the leak is due to an open fracture (porosity of 1) and 4 h of well pressurization during which well pressures are maintained at 7000 m of water head (approximately 1000 psi) (Adachi et al 2007;Myers 2012;Pitkin et al 2012;Kissinger et al 2013;NYDEC 2015), the minimum K required for transport over distances of 100, 500, and 2000 m during the period of hydraulic fracturing is estimated to be 9.9 × 10 −5 , 2.5 × 10 −3 , and 4.0 × 10 −2 m/s, respectively (Equation A1).…”

“…Many coal seams and shale aquifers are over-pressurized; over-pressures of 0.02 to 0.8 m/m are documented (Myers 2012;Gassiat et al 2013;Cai and Ofterdinger 2014), with an upward hydraulic gradient between the gas-bearing formation and any potential shallow aquifers. Although during development pressure in the gas-bearing formation is reduced, after gas production ceases, pressure will increase towards its initial value, but this recovery may take several hundred years (Cook et al 2017).…”